Generative AI on mobile and web with Google AI Edge

Unlock the secrets of bringing cutting-edge generative AI directly to your users' devices. We'll share how to seamlessly deploy popular gen AI models using the intuitive MediaPipe Tasks API. Plus, you'll discover how to customize and port your own gen AI creations specifically for edge devices using a powerful Python authoring API. Finally, learn how the new PyTorch to TensorFlow Lite converter can open a world of possibilities for on-device ML powered by PyTorch models.
AI Edge Model Explorer in Colab → goo.gle/model-explorer-colab
Model Explorer → goo.gle/ai-edge-model-explorer
Google AI Edge → goo.gle/ai-edge
LLM Inference → goo.gle/ai-edge-llm-api
AI models optimized and validated by Qualcomm Technologies → goo.gle/qc-ai-hub
Convert PyTorch models to TF Lite → goo.gle/ai-edge-torch
Convert generative PyTorch models → goo.gle/ai-edge-generative
Speakers: Aaron Karp, Cormac Brick, Sachin Kotwani
Watch more:
Check out all the AI videos at Google I/O 2024 → goo.gle/io24-ai-yt
Check out all the Mobile videos at Google I/O 2024 → goo.gle/io24-mobile-yt
Subscribe to Google Developers → goo.gle/developers
#GoogleIO
Event: Google I/O 2024

[@GoogleDevelopers]

Check out all the AI videos at Google I/O 2024 → goo.gle/io24-ai-yt

[@banzai316]

Yes Sir! 👏

[@rodrigolimadev]

i with to simply have a good example of a ML vision app using a Tflite exported from pytorch that has at least 40% of the accuracy of pytorch
that's all i wanted